Am I charging my client too much for my SLA - Service level agreement?

Question

So I have secured a project worth 6 figures for an employee management system that will manage his employees for 4 of his companies. This employee management system will integrate a biometric hardware solution on computer stations that communicate with iPads (for the managers/receptionist of each company) all connected to a cloud based server. The project will last around 3 months, and the customer is entitled to as many small intricate software changes as they like in the first year after which the service costs £8,000 a year to maintain the solution I'm about to implement.

Problem: Thats all great, but I'm having a hard time trying to explain I will be charging them 8k a year to keep things running after the first free year.

Here are my reasons for charging that:

Main two reasons

• I Blaaaady worked so hard in that first free year implementing all the feature changes that they suddenly realise they require new quick to implement, small feature upgrades sometimes big in the current software implementation that I feel I need to make my money back on the second and third year because of the extra hours put in after project completion.
• Apple no matter what will have Yearly iOS operating system upgrades for all devices, and so will the MAC OS X. So these applications will have to be updated regardless yearly anyway.
• Back ups of the server, and databases. <- this sounds really silly when I tell the client this.

Small extra reasons but that don't necessarily weigh significantly on the SLA cost

• I provide the client direct access to me, 24/7 telephone service as opposed to emails or any other communication method or having to go through my receptionist, I save that for the smaller projects. My client loves this and sure does use it whenever possible by dropping me many texts whenever a new feature is required, or when he needs something to change in the back end.
• Server costs, this is such a nominal amount that I could just pay this out of my own pocket since it only costs about £40 a month, not that I would take up the cost ;)
• Flexibility - we provide a no problem service attitude.

Everything is run on my side, if I wanted to, I just have to redirect the server elsewhere and all applications will cease to exist for the other projects I've worked on with this client.

But as far this goes, that's it. My client is asking me what will happen after the first free year if no updates are required in the second year, will he still need t pay the 8k? It's a good question, but that's not the point, I've told him these are the running costs for managing the service that I am providing.

Could someone please help me on outlining other important things that I am possibly missing from this SLA agreement that could help justify the cost? I find the app development aspect of it is more than enough to justify that cost, since some companies charge £25-30k in upgrades and app revisions where as he is able to come to me at any time of the year and send me a new list of features required with no hassle.

Answer 1

For SLA agreements you should be able to make an estimate based on fixed tasks and average expected work for improvements/bugs.

For fixed tasks:

• Server costs. If you're paying for this-- add it up and add this to the rest of the total. Don't just absorb it.
• Outline an update plan. For instance, Update all server software monthly. All device software monthly. Even if it ends up being trivial, expect it to be an X hour task you plan for and total up for the year. This makes you send organized at the very least.
• Backups and database -- if it is actually needed and you're doing it on a schedule, it counts. Describe the schedule. "Monthly directed backup of code and database. Recovery plan in case of fail.". If automated, probably a lesser verification+maintenance. Average the hours per month here and add in. Apply hourly rate.

For updates/fixes:

• Generally a combination of the average for past year and expectations for upcoming year if it will be dramatically slower. Have no updates been needed in the 2nd half of the first free year? You'll need to gauge this-- and assume a certain minimum here. If you plan for 1 fix for the year and you end up with 6, you're eating the cost. Better to have a minimum of 5 expected, and okay to eat a couple if happens. You're also being paid for being available for those cases, where otherwise you could commit your time exclusively to another client. It's a retainer of sorts. Estimate hours and apply hourly rate.
• 24/7 support. You mentioned it was for a smaller client, so this is probably minimal. I would note it though as part of the total you're trying to justify. You're providing something big here.
• Flexibility - Some companies like to see SLA expressed like "for moderate size issues: will acknowledge all issue requests via email within 4 hours, will propose solution within 8 hours, and will deploy solution within 48 hours.", "for significant issues, within 4 days". etc. These are ways to quantify the SLA service, and that you generally fix things fast. This is paid for, as the cheaper version would be the 1-2 week turnaround, etc.

For possibility where no fixes are needed, the SLA still applies. You're being paid for retaining possible availability, with fast/reliable turnaround. It's a guarantee to the client that if things go wrong, they'll be resolved. It's a 1 year contract, and perhaps the following year would be a reduced estimate based on the lower expectation of needs.

Answer 2

I don't think those are small reasons, especially the first one. Here's how I would explain it:

"I have to set aside time in my schedule to be available if you have questions or problems. That time costs money. If I don't set aside that time and you have an issue, suddenly all my other work has to be dropped by the way-side which is not fair to those other clients. I have to do this regardless of whether or not you use it."

Also, I don't think you sound silly at all regarding backups of the server and databases. Consider that this isn't just the backups, it is also the recovery if there is a problem. That will take your time, and it will happen at the worst possible moment. Your fee has to take into account that possibility. And let's face it, it isn't just an automated process to run the backup. You have to also occasionally verify that the backups are still happening (how many times have we seen something break those?)

Answer 3

Here's a different approach. Instead of 8k a year, offer the client your available time in smaller chunks:

8k / year 750 / mo 200 / seven business days 75 / per EACH incident (with some limitation!!!)

The client will have to get wise real quick. Backups and maintenance is nothing trivial, and maybe you set a separate price for just that alone.